-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
Hi List,
I've been working with a couple V240s, and have run into a few issues. I have tried 3 different machines, so I think I've fairly ruled out a hardware issue. With that said, here's what I see.
On both 2.4 and 2.6 kernels, I get an occasional D-cache parity error (no particular order, the Mar 4 ones are on a different machine than the later ones): Mar 4 18:34:08 v240 CPU[1]: Cheetah+ D-cache parity error at TPC[00000000004a62d8] Mar 4 10:50:07 v240 CPU[0]: Cheetah+ D-cache parity error at TPC[00000000004606c0] Mar 4 12:22:54 v240 CPU[1]: Cheetah+ D-cache parity error at TPC[00000000005ae90c] Mar 7 17:25:15 v240 CPU[0]: Cheetah+ D-cache parity error at TPC[00000000004870a4] Mar 8 02:08:22 v240 CPU[0]: Cheetah+ D-cache parity error at TPC[00000000004870a4] Mar 8 09:13:51 v240 CPU[0]: Cheetah+ D-cache parity error at TPC[00000000004870a4] Mar 10 04:53:45 v240 CPU[0]: Cheetah+ D-cache parity error at TPC[00000000004ec38c] Mar 11 02:33:32 v240 CPU[0]: Cheetah+ D-cache parity error at TPC[00000000004ec38c]
Since it is a cache error, it should be No Big Deal, and maybe Solaris is silently ignoring them?
These machines all have 2G of ram, but Linux detects otherwise. 2.4.29 says: Memory: 1293136k available (1712k kernel code, 240k data, 128k init) [fffff80000000000,000000103ff02000] 2.6.9 says: Memory: 1623296k available (1920k kernel code, 544k data, 144k init) [fffff80000000000,000000103ff02000]
I enabled debugging Bootmem on 2.4, it says: Bootmem_init: Scan sp_banks, init_bootmem(min[0], bootmap[162], max[81ff81]) free_bootmem(sp_banks:0): base[0] size[40000000] free_bootmem(sp_banks:1): base[1000000000] size[3effe000] free_bootmem(sp_banks:2): base[103f000000] size[dcc000] free_bootmem(sp_banks:3): base[103fdd0000] size[e0000] free_bootmem(sp_banks:4): base[103fec0000] size[10000] free_bootmem(sp_banks:5): base[103ff00000] size[2000] reserve_bootmem(kernel): base[0] size[2c3a90] reserve_bootmem(bootmap): base[2c4000] size[103ff8]
Anyway, I can live with that. However, after a period of Real Work, Mar 4 13:31:44 v240 ERROR(0): Cheetah error trap taken afsr[8010100000000000] afar[000000009af900e0] TL1(1) Mar 4 13:31:44 v240 ERROR(0): TPC[0000000000408d24] TNPC[0000000000408d28] TSTATE[0000009911049406] Mar 4 13:31:44 v240 ERROR(0): M_SYND(0), E_SYND(0), Privileged Mar 4 13:31:44 v240 ERROR(0): Highest priority error (0000100000000000) "Unmapped error from system bus" Mar 4 13:31:44 v240 ERROR(0): D-cache idx[80e0] tag[000000000009af91] utag[0000000000003300] stag[000000000009af90] Mar 4 13:31:44 v240 ERROR(0): D-cache data0[00135f2070178a14] data1[004b208c004b1f44] data2[0000000000030419] data3[0000000000030419] Mar 4 13:31:44 v240 ERROR(0): I-cache idx[0] tag[0000000000000000] utag[0000000000000000] stag[0000000000000000] u[0000000000000000] l[0000000000000000] Mar 4 13:31:44 v240 ERROR(0): I-cache INSN0[0000000000000000] INSN1[0000000000000000] INSN2[0000000000000000] INSN3[0000000000000000] Mar 4 13:31:44 v240 ERROR(0): I-cache INSN4[0000000000000000] INSN5[0000000000000000] INSN6[0000000000000000] INSN7[0000000000000000] Mar 4 13:31:44 v240 ERROR(0): E-cache idx[9af900e0] tag[0010100000000000] Mar 4 13:31:44 v240 ERROR(0): E-cache data0[106ff8c18f414000] data1[9210001494100015] data2[40002c549003a8bf] data3[306ff9da86102018] Mar 4 13:31:44 v240 Kernel panic: Irrecoverable deferred error trap. Mar 4 13:31:44 v240 Mar 4 13:31:45 v240 Press L1-A to return to the boot prom
When this happens, the box is still Alive, but one cpu is pegged at 100%. This is from Machine1. However, Machine2 agrees: Mar 5 11:16:34 v240 ERROR(0): Cheetah error trap taken afsr[8010800000000000] afar[00000009eb6340e0] TL1(1) Mar 5 11:16:34 v240 ERROR(0): TPC[0000000000410514] TNPC[0000000000410518] TSTATE[0000004411009507] Mar 5 11:16:34 v240 ERROR(0): M_SYND(0), E_SYND(0), Privileged Mar 5 11:16:34 v240 ERROR(0): Highest priority error (0000800000000000) "Out of range memory error has occurred" Mar 5 11:16:34 v240 ERROR(0): AFAR M-syndrome [???] Mar 5 11:16:34 v240 ERROR(0): D-cache idx[80e0] tag[00000000209eb635] utag[000000000000be00] stag[00000000209eb634] Mar 5 11:16:34 v240 ERROR(0): D-cache data0[413d6c696278736c] data1[0000000000414000] data2[fffff80030fd0118] data3[0000000000000000] Mar 5 11:16:34 v240 ERROR(0): I-cache idx[0] tag[0000000000000000] utag[0000000000000000] stag[0000000000000000] u[0000000000000000] l[0000000000000000] Mar 5 11:16:34 v240 ERROR(0): I-cache INSN0[0000000000000000] INSN1[0000000000000000] INSN2[0000000000000000] INSN3[0000000000000000] Mar 5 11:16:34 v240 ERROR(0): I-cache INSN4[0000000000000000] INSN5[0000000000000000] INSN6[0000000000000000] INSN7[0000000000000000] Mar 5 11:16:34 v240 ERROR(0): E-cache idx[eb6340e0] tag[0000000000000000] Mar 5 11:16:34 v240 ERROR(0): E-cache data0[0000000000000000] data1[0000000000000000] data2[0000000000000000] data3[0000000000000000] Mar 5 11:16:34 v240 Kernel panic: Irrecoverable deferred error trap. Mar 5 11:16:34 v240 Mar 5 11:16:35 v240 Press L1-A to return to the boot prom
These logs are from kernel 2.4.29. I also tried 2.6, and it hangs around when it would be starting init if I run SMP, boots ok UP.. Although if I force init=/bin/bash, it works. I've added echos into each file in the boot runlevel, and none of them ever print.
I ran a UP 2.6.9 kernel for over 4 days and got only D-cache errors, no crash.
When the first box crashes, I figured it was bad hardware. Since Sun would want to hear about how Solaris crashed it, I put Solaris 10 on it and ran SunVTS 6 in exclusive mode on the cpu and memory for a solid week, and it passed all tests. So, here I am.
Below I'm pasting a full dmesg from machine 1.
Any thoughts?
Thanks,
Josh
Mar 4 09:26:11 v240 PROMLIB: Sun IEEE Boot Prom 4.16.4 2004/12/18 05:20 Mar 4 09:26:11 v240 Linux version 2.4.29-sparc ([EMAIL PROTECTED]) (gcc version 3.3.5 (Gentoo Linux 3.3.5)) #1 SMP Thu Mar 3 13:06:23 CST 2005 Mar 4 09:26:11 v240 ARCH: SUN4U Mar 4 09:26:11 v240 Ethernet address: 00:03:ba:5f:9c:11 Mar 4 09:26:11 v240 On node 0 totalpages: 261498 Mar 4 09:26:11 v240 zone(0): 8519553 pages. Mar 4 09:26:11 v240 zone(1): 0 pages. Mar 4 09:26:11 v240 zone(2): 0 pages. Mar 4 09:26:11 v240 Found CPU 0 (node=f0065570,mid=0) Mar 4 09:26:11 v240 Found CPU 1 (node=f0065df0,mid=1) Mar 4 09:26:11 v240 Found 2 CPU prom device tree node(s). Mar 4 09:26:11 v240 Kernel command line: root=/dev/sda4 Mar 4 09:26:11 v240 Calibrating delay loop... 851.96 BogoMIPS Mar 4 09:26:11 v240 Memory: 1293136k available (1712k kernel code, 240k data, 128k init) [fffff80000000000,000000103ff02000] Mar 4 09:26:11 v240 Dentry cache hash table entries: 262144 (order: 9, 4194304 bytes) Mar 4 09:26:11 v240 Inode cache hash table entries: 131072 (order: 8, 2097152 bytes) Mar 4 09:26:11 v240 Mount cache hash table entries: 512 (order: 0, 8192 bytes) Mar 4 09:26:11 v240 Buffer cache hash table entries: 131072 (order: 7, 1048576 bytes) Mar 4 09:26:11 v240 Page-cache hash table entries: 262144 (order: 8, 2097152 bytes) Mar 4 09:26:11 v240 POSIX conformance testing by UNIFIX Mar 4 09:26:11 v240 Entering UltraSMPenguin Mode... Mar 4 09:26:11 v240 Calibrating delay loop... 851.96 BogoMIPS Mar 4 09:26:11 v240 Total of 2 processors activated (1703.93 BogoMIPS). Mar 4 09:26:11 v240 CPU 0: synchronized TICK with master CPU (last diff 1 cycles,maxerr 6 cycles) Mar 4 09:26:11 v240 Waiting on wait_init_idle (map = 0x1) Mar 4 09:26:11 v240 All processors have done init_idle Mar 4 09:26:11 v240 PCI: Probing for controllers. Mar 4 09:26:11 v240 TOMATILLO0 PBMB: ver[4:0], portid 1f, cregs[4000fc00000] pregs[4000ff00000] Mar 4 09:26:11 v240 TOMATILLO0 PBMB: PCI CFG[7f600000000] IO[7f601000000] MEM[7f700000000] Mar 4 09:26:11 v240 TOMATILLO0 PBMA: ver[4:0], portid 1e, cregs[4000f400000] pregs[4000f600000] Mar 4 09:26:11 v240 TOMATILLO0 PBMA: PCI CFG[7fe00000000] IO[7fe01000000] MEM[7ff00000000] Mar 4 09:26:11 v240 TOMATILLO1 PBMA: ver[4:0], portid 1c, cregs[4000e400000] pregs[4000e600000] Mar 4 09:26:11 v240 TOMATILLO1 PBMA: PCI CFG[7ce00000000] IO[7ce01000000] MEM[7cf00000000] Mar 4 09:26:11 v240 TOMATILLO1 PBMB: ver[4:0], portid 1d, cregs[4000ec00000] pregs[4000ef00000] Mar 4 09:26:11 v240 TOMATILLO1 PBMB: PCI CFG[7c600000000] IO[7c601000000] MEM[7c700000000] Mar 4 09:26:11 v240 PCI-IRQ: Routing bus[ 0] slot[ 2] map[0] to INO[1c] Mar 4 09:26:11 v240 PCI-IRQ: Routing bus[ 0] slot[ 2] map[0] to INO[1d] Mar 4 09:26:11 v240 PCI1(PBMB): Bus running at 66MHz Mar 4 09:26:11 v240 PCI-IRQ: Routing bus[ 1] slot[ 2] map[0] to INO[29] Mar 4 09:26:11 v240 PCI-IRQ: Routing bus[ 1] slot[ 2] map[0] to INO[28] Mar 4 09:26:11 v240 PCI1(PBMA): Bus running at 66MHz Mar 4 09:26:11 v240 PCI-IRQ: Routing bus[ 2] slot[ 2] map[0] to INO[08] Mar 4 09:26:11 v240 PCI-IRQ: Routing bus[ 2] slot[ 2] map[0] to INO[09] Mar 4 09:26:11 v240 PCI0(PBMB): Bus running at 66MHz Mar 4 09:26:11 v240 PCI-IRQ: Routing bus[ 3] slot[ a] map[0] to INO[27] Mar 4 09:26:11 v240 PCI-IRQ: Routing bus[ 3] slot[ d] map[0] to INO[18] Mar 4 09:26:11 v240 PCI0(PBMA): Bus running at 33MHz Mar 4 09:26:11 v240 isa0: [flashprom] [rtc] [i2c -> (i2c-bridge) (i2c-bridge) (motherboard-fru-prom) (chassis-fru-prom) (power-supply-fru-prom) (power-supply-fru-prom) (dimm-spd) (dimm-spd) (dimm-spd) (dimm-spd) (rscrtc) (nvram) (idprom) (gpio) (gpio) (gpio) (gpio) (gpio) (gpio)] [power] [serial] [serial] [rmc-comm] Mar 4 09:26:11 v240 ebus: No EBus's found. Mar 4 09:26:11 v240 PCIO serial driver version 1.54 Mar 4 09:26:11 v240 su(serial) at 0x7fe010003f8 (tty 0 irq 4,7ac) is a 16550A Mar 4 09:26:11 v240 su(serial) at 0x7fe010002e8 (tty 1 irq 4,7ac) is a 16550A Mar 4 09:26:11 v240 Linux NET4.0 for Linux 2.4 Mar 4 09:26:11 v240 Based upon Swansea University Computer Society NET3.039 Mar 4 09:26:11 v240 Initializing RT netlink socket Mar 4 09:26:11 v240 Starting kswapd Mar 4 09:26:11 v240 Journalled Block Device driver loaded Mar 4 09:26:11 v240 devfs: v1.12c (20020818) Richard Gooch ([EMAIL PROTECTED]) Mar 4 09:26:11 v240 devfs: boot_options: 0x1 Mar 4 09:26:11 v240 i2c-core.o: i2c core module version 2.6.1 (20010830) Mar 4 09:26:11 v240 i2c-dev.o: i2c /dev entries driver module version 2.6.1 (20010830) Mar 4 09:26:11 v240 i2c-core.o: driver i2c-dev dummy driver registered. Mar 4 09:26:11 v240 i2c-algo-bit.o: i2c bit algorithm module Mar 4 09:26:11 v240 i2c-proc.o version 2.6.1 (20010830) Mar 4 09:26:11 v240 pty: 256 Unix98 ptys configured Mar 4 09:26:11 v240 Real Time Clock Driver v1.10f Mar 4 09:26:11 v240 tg3.c:v3.15 (January 6, 2005) Mar 4 09:26:11 v240 eth0: Tigon3 [partno(Sun 570X) rev 2003 PHY(5704)] (PCI:66MHz:64-bit) 10/100/1000BaseT Ethernet 00:03:ba:5f:9c:13 Mar 4 09:26:11 v240 eth0: RXcsums[1] LinkChgREG[0] MIirq[0] ASF[0] Split[0] WireSpeed[1] TSOcap[0] Mar 4 09:26:11 v240 eth1: Tigon3 [partno(Sun 570X) rev 2003 PHY(5704)] (PCI:66MHz:64-bit) 10/100/1000BaseT Ethernet 00:03:ba:5f:9c:14 Mar 4 09:26:11 v240 eth1: RXcsums[1] LinkChgREG[0] MIirq[0] ASF[0] Split[0] WireSpeed[1] TSOcap[0] Mar 4 09:26:11 v240 eth2: Tigon3 [partno(Sun 570X) rev 2003 PHY(5704)] (PCI:66MHz:64-bit) 10/100/1000BaseT Ethernet 00:03:ba:5f:9c:11 Mar 4 09:26:11 v240 eth2: RXcsums[1] LinkChgREG[0] MIirq[0] ASF[0] Split[0] WireSpeed[1] TSOcap[0] Mar 4 09:26:11 v240 eth3: Tigon3 [partno(Sun 570X) rev 2003 PHY(5704)] (PCI:66MHz:64-bit) 10/100/1000BaseT Ethernet 00:03:ba:5f:9c:12 Mar 4 09:26:11 v240 eth3: RXcsums[1] LinkChgREG[0] MIirq[0] ASF[0] Split[0] WireSpeed[1] TSOcap[0] Mar 4 09:26:11 v240 SCSI subsystem driver Revision: 1.00 Mar 4 09:26:11 v240 sym.1.2.0: setting PCI_COMMAND_INVALIDATE. Mar 4 09:26:11 v240 sym.1.2.1: setting PCI_COMMAND_INVALIDATE. Mar 4 09:26:11 v240 sym0: <1010-66> rev 0x1 on pci bus 1 device 2 function 0 irq 4,729 Mar 4 09:26:11 v240 sym0: No NVRAM, ID 7, Fast-80, LVD, parity checking Mar 4 09:26:11 v240 sym0: SCSI BUS has been reset. Mar 4 09:26:11 v240 sym1: <1010-66> rev 0x1 on pci bus 1 device 2 function 1 irq 4,728 Mar 4 09:26:11 v240 sym1: No NVRAM, ID 7, Fast-80, LVD, parity checking Mar 4 09:26:11 v240 sym1: SCSI BUS has been reset. Mar 4 09:26:11 v240 scsi0 : sym-2.1.17a Mar 4 09:26:11 v240 scsi1 : sym-2.1.17a Mar 4 09:26:11 v240 Vendor: FUJITSU Model: MAP3367N SUN36G Rev: 0401 Mar 4 09:26:11 v240 Type: Direct-Access ANSI SCSI revision: 04 Mar 4 09:26:11 v240 Vendor: FUJITSU Model: MAP3367N SUN36G Rev: 0401 Mar 4 09:26:11 v240 Type: Direct-Access ANSI SCSI revision: 04 Mar 4 09:26:11 v240 sym0:0:0: tagged command queuing enabled, command queue depth 16. Mar 4 09:26:11 v240 sym0:1:0: tagged command queuing enabled, command queue depth 16. Mar 4 09:26:11 v240 Attached scsi disk sda at scsi0, channel 0, id 0, lun 0 Mar 4 09:26:11 v240 Attached scsi disk sdb at scsi0, channel 0, id 1, lun 0 Mar 4 09:26:11 v240 sym0:0: FAST-40 WIDE SCSI 80.0 MB/s ST (25.0 ns, offset 31) Mar 4 09:26:11 v240 SCSI device sda: 71132959 512-byte hdwr sectors (36420 MB) Mar 4 09:26:11 v240 Partition check: Mar 4 09:26:11 v240 /dev/scsi/host0/bus0/target0/lun0: p1 p2 p3 p4 Mar 4 09:26:11 v240 sym0:1: FAST-40 WIDE SCSI 80.0 MB/s ST (25.0 ns, offset 31) Mar 4 09:26:11 v240 SCSI device sdb: 71132959 512-byte hdwr sectors (36420 MB) Mar 4 09:26:11 v240 /dev/scsi/host0/bus0/target1/lun0: p1 p3 Mar 4 09:26:11 v240 NET4: Linux TCP/IP 1.0 for NET4.0 Mar 4 09:26:11 v240 IP Protocols: ICMP, UDP, TCP, IGMP Mar 4 09:26:11 v240 IP: routing cache hash table of 16384 buckets, 256Kbytes Mar 4 09:26:11 v240 TCP: Hash tables configured (established 262144 bind 65536) Mar 4 09:26:11 v240 NET4: Unix domain sockets 1.0/SMP for Linux NET4.0. Mar 4 09:26:11 v240 kjournald starting. Commit interval 5 seconds Mar 4 09:26:11 v240 EXT3-fs: mounted filesystem with ordered data mode. Mar 4 09:26:11 v240 VFS: Mounted root (ext3 filesystem) readonly. Mar 4 09:26:11 v240 Mounted devfs on /dev Mar 4 09:26:11 v240 Adding Swap: 499776k swap-space (priority -1) Mar 4 09:26:11 v240 EXT3 FS 2.4-0.9.19, 19 August 2002 on sd(8,4), internal journal Mar 4 09:26:11 v240 kjournald starting. Commit interval 5 seconds Mar 4 09:26:11 v240 EXT3 FS 2.4-0.9.19, 19 August 2002 on sd(8,17), internal journal Mar 4 09:26:11 v240 EXT3-fs: mounted filesystem with ordered data mode. Mar 4 09:26:11 v240 TOMATILLO1 PBMB: PCI Error, primary error type[Excessive Retries] Mar 4 09:26:11 v240 TOMATILLO1 PBMB: bytemask[0000] was_block(0) space(Config) Mar 4 09:26:11 v240 TOMATILLO1 PBMB: PCI AFAR [0000000000002084] Mar 4 09:26:11 v240 TOMATILLO1 PBMB: PCI Secondary errors [(none)] Mar 4 09:26:11 v240 tg3: eth0: Link is up at 100 Mbps, half duplex. Mar 4 09:26:11 v240 tg3: eth0: Flow control is off for TX and off for RX. Mar 4 09:26:11 v240 envctrl: I2C device not found. Mar 4 09:26:11 v240 kjournald starting. Commit interval 5 seconds Mar 4 09:26:11 v240 EXT3 FS 2.4-0.9.19, 19 August 2002 on sd(8,1), internal journal Mar 4 09:26:11 v240 EXT3-fs: mounted filesystem with ordered data mode. Mar 4 10:50:07 v240 CPU[0]: Cheetah+ D-cache parity error at TPC[00000000004606c0] Mar 4 12:22:54 v240 CPU[1]: Cheetah+ D-cache parity error at TPC[00000000005ae90c] Mar 4 13:31:44 v240 ERROR(0): Cheetah error trap taken afsr[8010100000000000] afar[000000009af900e0] TL1(1) Mar 4 13:31:44 v240 ERROR(0): TPC[0000000000408d24] TNPC[0000000000408d28] TSTATE[0000009911049406] Mar 4 13:31:44 v240 ERROR(0): M_SYND(0), E_SYND(0), Privileged Mar 4 13:31:44 v240 ERROR(0): Highest priority error (0000100000000000) "Unmapped error from system bus" Mar 4 13:31:44 v240 ERROR(0): D-cache idx[80e0] tag[000000000009af91] utag[0000000000003300] stag[000000000009af90] Mar 4 13:31:44 v240 ERROR(0): D-cache data0[00135f2070178a14] data1[004b208c004b1f44] data2[0000000000030419] data3[0000000000030419] Mar 4 13:31:44 v240 ERROR(0): I-cache idx[0] tag[0000000000000000] utag[0000000000000000] stag[0000000000000000] u[0000000000000000] l[0000000000000000] Mar 4 13:31:44 v240 ERROR(0): I-cache INSN0[0000000000000000] INSN1[0000000000000000] INSN2[0000000000000000] INSN3[0000000000000000] Mar 4 13:31:44 v240 ERROR(0): I-cache INSN4[0000000000000000] INSN5[0000000000000000] INSN6[0000000000000000] INSN7[0000000000000000] Mar 4 13:31:44 v240 ERROR(0): E-cache idx[9af900e0] tag[0010100000000000] Mar 4 13:31:44 v240 ERROR(0): E-cache data0[106ff8c18f414000] data1[9210001494100015] data2[40002c549003a8bf] data3[306ff9da86102018] Mar 4 13:31:44 v240 Kernel panic: Irrecoverable deferred error trap. Mar 4 13:31:44 v240 Mar 4 13:31:45 v240 Press L1-A to return to the boot prom -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.6 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org
iD8DBQFCMe88FAhB33r2ACYRAjcJAJ9WH/gSvsNZVgPQyDy+cfeGHWfcPwCcCVzz tmKo+pmumZNVS2mrRsiITFs= =iORx -----END PGP SIGNATURE----- - To unsubscribe from this list: send the line "unsubscribe sparclinux" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html
